H. S. Williams — Crustacean from the Devonian. 45 



pansion of the glass. The amount for the interval 0°-100° is 

 readily calculated, and presents us with the following values : 



T. 0. 81 100. (5/3 = — 0-000046 



Y. 0. 89 = — 0000022 



Y. 0. 90 = — 0-000003 



Y. 0. 20967.. = + 0-000026 



Yalues similar to the first two have been previously ob- 

 served by Weber and by Crafts. The latter two are, as far as 

 my knowledge goes, without precedent. 



Again, if the thermometer be preserved at ordinary tempera- 

 tures, similar changes occur. We may instance the elevation 

 of the zero with time. Its law is similar to that of the eleva- 

 tion produced by heating ; or, analogies occur in the change 

 in the fundamental length and the indicated change in the 

 coefficient of expansion. 



3. 



200= 

 (Fahrenheit,,) 



Difference in the correction before and after treatment. 



All of these facts — the condensing of the material forming 

 the bulb, the consequent increase of its intermolecular attraction, 

 the dependence of the point of rising upon its chemical consti- 

 tution, the similarity of the changes produced by time, the 

 regularity of these changes — seem to indicate the cause for the 

 one as was long since suggested for the other, in a partial sepa- 

 ration of the crystalline from the amorphous bulb-constituents. 

 If the view is correct, it argues well for the stability of the 

 treated instrument. The change will have been produced at 

 the expense of its natural life. But then, few thermometers 

 are permitted to die of old age. The correctness of the view 

 is the subject of a separate research. 



Yale College Observatory, April, 1885. 



Art. VII. — Notice of a new Limuloid Crustacean from the 

 Devonian ; by Henry Shaler Williams. 



Among the fossils collected last summer for a comparative 

 study of the Devonian faunas, an interesting form was dis- 

 covered in Erie County, Pennsylvania, worthy of special notice. 



The specimen was found in the bluish sandstone (which in 

 places is a fine pebbly conglomerate) at Le Bceuf, called the 

 "3d oil sand'' by Mr. I. C. White in the Report Q 4 of the 

 Second Geological Survey of Pennsylvania (p. 239), and re- 



